CN107985200A - A kind of load truck right-hand bend safety pre-warning system and method - Google Patents
A kind of load truck right-hand bend safety pre-warning system and method Download PDFInfo
- Publication number
- CN107985200A CN107985200A CN201711112829.0A CN201711112829A CN107985200A CN 107985200 A CN107985200 A CN 107985200A CN 201711112829 A CN201711112829 A CN 201711112829A CN 107985200 A CN107985200 A CN 107985200A
- Authority
- CN
- China
- Prior art keywords
- pedestrian
- fish
- hand bend
- load truck
- blind area
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60R—VEHICLES, VEHICLE FITTINGS, OR VEHICLE PARTS, NOT OTHERWISE PROVIDED FOR
- B60R1/00—Optical viewing arrangements; Real-time viewing arrangements for drivers or passengers using optical image capturing systems, e.g. cameras or video systems specially adapted for use in or on vehicles
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60R—VEHICLES, VEHICLE FITTINGS, OR VEHICLE PARTS, NOT OTHERWISE PROVIDED FOR
- B60R16/00—Electric or fluid circuits specially adapted for vehicles and not otherwise provided for; Arrangement of elements of electric or fluid circuits specially adapted for vehicles and not otherwise provided for
- B60R16/02—Electric or fluid circuits specially adapted for vehicles and not otherwise provided for; Arrangement of elements of electric or fluid circuits specially adapted for vehicles and not otherwise provided for electric constitutive elements
- B60R16/023—Electric or fluid circuits specially adapted for vehicles and not otherwise provided for; Arrangement of elements of electric or fluid circuits specially adapted for vehicles and not otherwise provided for electric constitutive elements for transmission of signals between vehicle parts or subsystems
- B60R16/0231—Circuits relating to the driving or the functioning of the vehicle
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60T—VEHICLE BRAKE CONTROL SYSTEMS OR PARTS THEREOF; BRAKE CONTROL SYSTEMS OR PARTS THEREOF, IN GENERAL; ARRANGEMENT OF BRAKING ELEMENTS ON VEHICLES IN GENERAL; PORTABLE DEVICES FOR PREVENTING UNWANTED MOVEMENT OF VEHICLES; VEHICLE MODIFICATIONS TO FACILITATE COOLING OF BRAKES
- B60T7/00—Brake-action initiating means
- B60T7/12—Brake-action initiating means for automatic initiation; for initiation not subject to will of driver or passenger
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60R—VEHICLES, VEHICLE FITTINGS, OR VEHICLE PARTS, NOT OTHERWISE PROVIDED FOR
- B60R2300/00—Details of viewing arrangements using cameras and displays, specially adapted for use in a vehicle
- B60R2300/10—Details of viewing arrangements using cameras and displays, specially adapted for use in a vehicle characterised by the type of camera system used
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60R—VEHICLES, VEHICLE FITTINGS, OR VEHICLE PARTS, NOT OTHERWISE PROVIDED FOR
- B60R2300/00—Details of viewing arrangements using cameras and displays, specially adapted for use in a vehicle
- B60R2300/20—Details of viewing arrangements using cameras and displays, specially adapted for use in a vehicle characterised by the type of display used
- B60R2300/202—Details of viewing arrangements using cameras and displays, specially adapted for use in a vehicle characterised by the type of display used displaying a blind spot scene on the vehicle part responsible for the blind spot
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60R—VEHICLES, VEHICLE FITTINGS, OR VEHICLE PARTS, NOT OTHERWISE PROVIDED FOR
- B60R2300/00—Details of viewing arrangements using cameras and displays, specially adapted for use in a vehicle
- B60R2300/30—Details of viewing arrangements using cameras and displays, specially adapted for use in a vehicle characterised by the type of image processing
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60R—VEHICLES, VEHICLE FITTINGS, OR VEHICLE PARTS, NOT OTHERWISE PROVIDED FOR
- B60R2300/00—Details of viewing arrangements using cameras and displays, specially adapted for use in a vehicle
- B60R2300/80—Details of viewing arrangements using cameras and displays, specially adapted for use in a vehicle characterised by the intended use of the viewing arrangement
- B60R2300/802—Details of viewing arrangements using cameras and displays, specially adapted for use in a vehicle characterised by the intended use of the viewing arrangement for monitoring and displaying vehicle exterior blind spot views
- B60R2300/8026—Details of viewing arrangements using cameras and displays, specially adapted for use in a vehicle characterised by the intended use of the viewing arrangement for monitoring and displaying vehicle exterior blind spot views in addition to a rear-view mirror system
Landscapes
- Engineering & Computer Science (AREA)
- Mechanical Engineering (AREA)
- Multimedia (AREA)
- Automation & Control Theory (AREA)
- Transportation (AREA)
- Traffic Control Systems (AREA)
Abstract
The invention discloses a kind of load truck right-hand bend safety pre-warning system and method based on flake vision.The system comprises fish-eye camera, ultrasonic sensor, signal processing unit, Central Control Center, liquid crystal display, throttle control.The method mainly comprises the following steps:When starting right turn signal, start fish-eye camera and ultrasonic sensor carries out the collection of load truck right-hand bend dead zone information.The calibration of fish-eye camera, correcting distorted image are carried out by disk scaling board.Load truck right-hand bend blind area pedestrian detection is carried out by HOG+SVM algorithms, then carries out load truck right-hand bend blind area pedestrian's real-time tracking using the correlation filtering blending algorithm based on color characteristic statistics and HOG character representations.The people's car lateral separation measured according to ultrasonic sensor, controls load truck turning speed, prevents driver's maloperation.
Description
Technical field
The present invention relates to vehicle drive security technology area, more particularly to load truck right-hand bend safety pre-warning system and side
Method.
Background technology
In recent years, since the needs of economic development and industrial transport, the ownership of China's load truck are continuously increased.Load-carrying
Lorry makes it result in car accident accordingly there are many vision dead zones and constantly occur because of the characteristics of its " big, long, high ".
Particularly " dead bent moon " accident, i.e., when load truck is turned right, because the presence of lubrication groove difference and right rear view mirror blind area, drives
Member is not it will be noted that whether there is pedestrian, so as to cause the generation of accident around load truck right side bodies.
The Chinese patent of patent publication No. CN104401254A discloses death domain human body sensing system when lorry is turned,
It is related to a kind of infrared induction system.The system includes multiple infrared inductors, is distributed in the both sides of lorry, connects goods respectively
Steering, whistle and the brake gear of car.Since lubrication groove difference has about 2 meters of death domain when lorry is turned, driver can not lead to
Later visor observes, and the sum of induction range of infrared inductor covering rectangle death domain, turns when lorry and senses system
For system when sensing human body, infrared inductor can send signal to whistle and brake gear, by blowing a whistle and brake avoids accident
Occur.Although the patent can effectively detect death domain pedestrian, system signal collection is relatively simple, if infrared inductor
Accidentally break down, system is by cisco unity malfunction.
The Chinese patent of patent publication No. CN104477094A discloses a kind of trailer turning security early warning system, belongs to friendship
Logical security monitoring field.The system includes vehicle body, and driver's cabin turns to signal controlled switch, pedestrian's cell phone terminal, and driver moves
Dynamic mobile phone terminal, is arranged on the information collecting device and alarming device of vehicle body both sides, and is arranged on and drives indoor alarm dress
Put and central processing unit.The system obtains signal come early warning driver and pedestrian using two-way, and signal is all the way:Adopted by information
Acquisition means gather traffic information and incoming central processing unit, after central processing unit carries out conversion and logical operation, output control
Signal is to alarming device and warning device.Another way signal is:Pedestrian's cell phone terminal and driver's cell phone terminal will
Location information communication is connected to central processing unit, carries out calculation process through central processing unit, output triggers driver and in danger
The cell phone terminal alarms signal of the pedestrian of dangerous distance.The system is used as letter using millimetre-wave radar and wide-angle camera at the same time
Harvester, LED soft lights band are ceased as alarming device, buzzer as warning device.Although this system overcomes signal acquisition
The shortcomings that single, the reliability of energy safeguards system, but the system equipment is excessive, and signal processing real-time is poor, and passes through mobile phone
Terminal is alarmed.If driver and pedestrian do not carry mobile terminal, alarm will fail.
The content of the invention
It is an object of the present invention to provide a kind of load truck right-hand bend safety pre-warning system.It can efficiently solve load
Whether heavy cargo car right-hand rotation blind area has the observation problem of the vision of obstacle.
It is a further object to provide a kind of load truck right-hand bend safe early warning method.It can be efficiently solved
There is the early warning problem of obstacle in load truck right-hand rotation blind area.
The purpose of the present invention is what is be achieved through the following technical solutions:A kind of load truck right-hand bend safe early warning system
System, including fish-eye camera, ultrasonic sensor, signal processing unit, Central Control Center, liquid crystal display, Throttle Opening Control
Device;
Bridge is equipped with liquid crystal display, and railway carriage or compartment back is equipped with fish-eye camera after truck cab, is set on the right side of boxcar
There is more than one ultrasonic sensor, cabin interior is equipped with signal processing unit and Central Control Center;Fish-eye camera
Output terminal the input terminal of signal processing unit, the output terminal connection of signal processing unit are connected with ultrasonic sensor output terminal
The input terminal of Central Control Center and the input terminal of liquid crystal display, the output terminal connection fish-eye camera of Central Control Center
The input terminal of input terminal, the input terminal of ultrasonic sensor and throttle control.
The fish-eye camera is arranged at railway carriage or compartment back after truck cab, and the ultrasonic sensor has multiple, difference position
Position on the right side of boxcar, the signal processing unit are arranged at cabin interior, and the Central Control Center is arranged at
Inside driver's cabin bridge, the liquid crystal display is arranged at driver's cabin bridge;
Another object of the present invention is achieved through the following technical solutions:A kind of load truck is turned right safe pre-
Alarm method, it includes the following steps:
Step 1: load truck right-hand bend dead zone information gathers:
After the Central Control Center of load truck detects right turn signal, the flake at railway carriage or compartment back is taken the photograph after startup truck cab
As head, real-time right-hand bend blind area video is obtained, while starts the ultrasonic sensor on the right side of load truck compartment, obtains distance letter
Breath;
Step 2: fish-eye camera is demarcated using disk scaling board:
Because fish eye lens shoot come photo it is approximate with disk, using one kind using disk scaling board offer accurately
Dotted line feature completes the calibration of fish-eye camera, passes through the inside and outside parameter and distortion factor demarcated and obtain fish-eye camera;
Step 3: the correction of flake fault image:
Fault image is corrected using the template in fish eye images function correction method, and with cubic convolution method to school
Image after just is filled, and is done homework for the detection and tracking of pedestrian in right-hand bend blind area;
Step 4: load truck right-hand bend blind area pedestrian detection and tracking:
After correcting distorted image, the detection and tracking processing of pedestrian are carried out in signal processing unit, chooses correction
The first two field picture afterwards carries out load truck right-hand bend blind area pedestrian detection;Using the character representation of histograms of oriented gradients HOG
Pedestrian detection is carried out plus the classifier algorithm of support vector machines;After detecting blind area pedestrian, the tracking box of the first frame is extracted
Positional information, blind area pedestrian tracking is carried out using the correlation filtering blending algorithm based on color characteristic and HOG character representations;
Step 5: load truck right-hand bend speed controls:
Load truck right-hand bend blind area pedestrian information is shown on liquid crystal display and driver can be helped to understand in real time
Dead zone information, but driver fault misstepping on accelerator in order to prevent, using a kind of load truck right-hand bend car based on people's car lateral separation
Fast control program, in fish-eye camera detecting and tracking blind area pedestrian at the same time, ultrasonic sensor obtain the transverse direction of pedestrian and lorry
Distance, Central Control Center is according to the lateral separation between current lorry turning speed and pedestrian, active control lorry throttle control
Device processed, prevents driver's maloperation;The specific method of control is:
Under any lateral separation, when throttle increases suddenly, Central Control Center carries out emergency braking;
A、10m<For lorry to the lateral separation of pedestrian, Central Control Center control throttle, limits turning speed 30km/h;
B、5m<Lateral separation of the lorry to pedestrian<10m, Central Control Center control throttle, limitation turning speed 20km/
h;
C、3m<Lateral separation of the lorry to pedestrian<5m, Central Control Center control throttle, limitation turning speed 10km/h;
D、1m<Lateral separation of the lorry to pedestrian<3m Central Control Center controls throttle, limitation turning speed 5km/h;
E, lateral separation of the lorry to pedestrian<1m, Central Control Center control throttle, limitation turning speed 0km/h;
Repeat the operation that above step one arrives step 5.
The present invention seeks to what is be achieved through the following technical solutions:
Fish-eye camera is demarcated using disk scaling board, which specifically includes:
B1, using fish-eye camera shoot multiple chequered with black and white disk scaling board images;
B2, carry out binaryzation, and cromogram is converted into gray-scale map;
B3, input skew disk uncalibrated image quantity, i.e., the number in the center of circle in transverse and longitudinal coordinate;
B4, using lower-left circle disk center as image coordinate origin, extract the position of each circle disk center as characteristic point;
B5, do sub-pix precision, keeps these centers of circle;
B6, according to camera coordinates system and world coordinate system transfer principle, calculate spin matrix in outer parameter and flat
The amount of shifting to, according to camera coordinates system and image physical coordinates system transfer principle, calculates intrinsic parameter;
B7, using the nonlinear model formula of camera calculate coefficient of radial distortion and tangential distortion coefficient.
Pedestrian detection is carried out to load truck right-hand bend blind area and tracking, the step specifically include:
D1, import the lorry blind area picture after the correction of n frames, chooses t frames;
D2, extraction area-of-interest RIO, PCA dimensionality reductions are carried out to ROI region;
Histograms of oriented gradients HOG features in D3, extraction test sample, utilize the support vector machines after training point
Class device carries out the identification of blind area pedestrian;
D4, discriminate whether to detect pedestrian;
If D5, it is unidentified arrive pedestrian, choose latter two field picture, carry out D2;
If D6, recognize pedestrian, tracking box (groundtruth) four corner location information are extracted;
D7, according to tracking box (groundtruth) positional information, the color characteristic training of target in frame is carried out, before statistics
The color probability in scape target and background region;
D8, according to tracking box (groundtruth) positional information, the HOG features of detection pedestrian are simultaneously instructed in extraction frame
Practice, obtain associated filter template;
Image after D9, input next frame correction;
D10, using bayes method differentiate that each pixel in image belongs to the probability of prospect, then suppresses the similar face in edge
The object of color, obtains target following region of the present frame based on color characteristic;
D11, do Cosine Window processing, is then multiplied by Fast Fourier Transform (FFT) with associated filter template, is Fourier
Inverse transformation obtains peak response point diagram, obtains the target following region that present frame is obtained based on correlation filtering;
D12, the peak response point diagram progress for obtaining the target following region counted based on color characteristic and correlation filtering
Fusion, obtains final tracking box (groundtruth) region of target in present frame, is then transferred to D7.
The beneficial effects of the invention are as follows:
Compared with prior art, the present invention obtains more lorry right-hand bend fade chart picture letters using fish-eye camera
Breath.By fish-eye calibration and the correction of fault image, using HOG+SVM pedestrian detections algorithm and based on color and
The pedestrian tracking algorithm of correlation filtering can timely and accurately detecting and tracking to the pedestrian in lorry right-hand rotation vision dead zone.After processing
Image display on liquid crystal display, alerting drivers can be played.In addition, driver is because seeing pedestrian in order to prevent
The maloperation in blind area, the present invention obtain the lateral separation between blind area pedestrian and lorry using ultrasonic sensor, according to
Distance and the operational circumstances of driver, speed when limitation lorry is turned.
In short, the present invention can when load truck is turned alerting drivers, turning speed can also be controlled, prevent from driving
The maloperation of member.System structure is simple, and replicability is strong.
Brief description of the drawings
Fig. 1 is the structure diagram of the present invention;
Fig. 2 is flow chart of the present invention;
The fish-eye camera that Fig. 3 is the present invention demarcates flow chart;
Fig. 4 is the load truck right-hand rotation blind area pedestrian detection and tracking flow chart of the present invention.
Embodiment
In order to make the purpose , technical scheme and advantage of the present invention be clearer, with reference to the accompanying drawings and embodiments, it is right
The present invention is further elaborated.
Implementation column 1:
Shown in please referring to Fig.1, Fig. 1 is the structural representation of 1 load truck right-hand bend safety pre-warning system of the embodiment of the present invention
Figure.A kind of load truck right-hand bend safety pre-warning system is present embodiments provided, the system comprises fish-eye camera 1, ultrasound
Wave sensor 2, signal processing unit 3, Central Control Center 4, liquid crystal display 5, throttle control 6;
It is latter half of that the fish-eye camera 1 is arranged at lorry headstock, and the ultrasonic sensor 2 has multiple, is located at respectively
Position on the right side of boxcar, the signal processing unit 3 are arranged at cabin interior, and the Central Control Center 4 is arranged at
Cabin interior, the liquid crystal display 5 are arranged at cabin interior;
Output terminal, the input terminal of 2 output terminal of ultrasonic sensor connection signal processing unit 3 of the fish-eye camera 1,
The input terminal of output terminal connection Central Control Center 4 and the input terminal of liquid crystal display 5 of the signal processing unit 3, it is described
Input terminal, the input terminal and Throttle Opening Control of ultrasonic sensor 2 of the output terminal connection fish-eye camera 1 of Central Control Center 4
The input terminal of device 6.
Implementation column 2:
Shown in please referring to Fig.2, Fig. 2 is load truck right-hand bend safety pre-warning system application method flow chart of the present invention.This
Embodiment provides the application method of the present invention, and this method comprises the following steps:
Step 1: load truck right-hand bend dead zone information gathers:
After the Central Control Center of load truck detects right turn signal, the flake at railway carriage or compartment back is taken the photograph after startup truck cab
As head, real-time right-hand bend blind area video is obtained, while starts the ultrasonic sensor on the right side of load truck compartment, obtains distance letter
Breath;
Step 2: fish-eye camera is demarcated using disk scaling board:
Because fish eye lens shoot come photo it is approximate with disk, using one kind using disk scaling board offer accurately
Dotted line feature completes the calibration of fish-eye camera, passes through the inside and outside parameter and distortion factor demarcated and obtain fish-eye camera;
Step 3: the correction of flake fault image:
Fault image is corrected using the template in fish eye images function correction method, and with cubic convolution method to school
Image after just is filled, and is done homework for the detection and tracking of pedestrian in right-hand bend blind area;
Step 4: load truck right-hand bend blind area pedestrian detection and tracking:
After correcting distorted image, the detection and tracking processing of pedestrian are carried out in signal processing unit, chooses correction
The first two field picture afterwards carries out load truck right-hand bend blind area pedestrian detection;Using the character representation of histograms of oriented gradients HOG
Pedestrian detection is carried out plus the classifier algorithm of support vector machines;After detecting blind area pedestrian, the tracking box of the first frame is extracted
Positional information, blind area pedestrian tracking is carried out using the correlation filtering blending algorithm based on color characteristic and HOG character representations;
Step 5: load truck right-hand bend speed controls:
Load truck right-hand bend blind area pedestrian information is shown on liquid crystal display and driver can be helped to understand in real time
Dead zone information, but driver fault misstepping on accelerator in order to prevent, using a kind of load truck right-hand bend car based on people's car lateral separation
Fast control program, in fish-eye camera detecting and tracking blind area pedestrian at the same time, ultrasonic sensor obtain the transverse direction of pedestrian and lorry
Distance, Central Control Center is according to the lateral separation between current lorry turning speed and pedestrian, active control lorry throttle control
Device processed, prevents driver's maloperation;The specific method of control is:
Under any lateral separation, when throttle increases suddenly, Central Control Center carries out emergency braking;
A、10m<For lorry to the lateral separation of pedestrian, Central Control Center control throttle, limits turning speed 30km/h;
B、5m<Lateral separation of the lorry to pedestrian<10m, Central Control Center control throttle, limitation turning speed 20km/
h;
C、3m<Lateral separation of the lorry to pedestrian<5m, Central Control Center control throttle, limitation turning speed 10km/h;
D、1m<Lateral separation of the lorry to pedestrian<3m Central Control Center controls throttle, limitation turning speed 5km/h;
E, lateral separation of the lorry to pedestrian<1m, Central Control Center control throttle, limitation turning speed 0km/h;
Repeat the operation that above step one arrives step 5.
Implementation column 3:
Shown in please referring to Fig.3, Fig. 3 demarcates flow chart for 3 fish-eye camera of the embodiment of the present invention.This method is in embodiment 2
Realized during step 2, this method comprises the following steps:
B1, using fish-eye camera shoot multiple chequered with black and white disk scaling board images.
B2, carry out binaryzation, and cromogram is converted into gray-scale map.
B3, input skew disk uncalibrated image quantity, i.e., the number in the center of circle in transverse and longitudinal coordinate.
B4, using lower-left circle disk center as image coordinate origin, extract the position (x of each circle disk center0,y0) as special
Sign point.
B5, do sub-pix precision, keeps these centers of circle.
B6, according to camera coordinates system midpoint (Xc, Yc, Zc) and world coordinate system midpoint (Xw, Yw, Zw) transfer principle:
Wherein 0TFor 0 transposed matrix.Calculate the spin matrix R and translation vector t in outer parameter.According to camera coordinates
It is midpoint (Xc, Yc, Zc) and image physical coordinates system midpoint (x, y) transfer principle:
S is a scale factor.The intrinsic parameter f of camera can be calculatedx,fy,cx,cy。
B7, the nonlinear model formula using camera:
Formula midpoint (xd,yd) it is home position, point (xp,yp) for correction after new position, r be a range Imaging instrument away from
From.Coefficient of radial distortion k can be calculated1,k2,k3,k4And tangential distortion coefficient p1,p2。
Implementation column 4:
Shown in please referring to Fig.4, Fig. 4 is 4 load truck right-hand rotation blind area pedestrian detection of the embodiment of the present invention and tracking stream
Cheng Tu.This method realizes that this method comprises the following steps during 2 step 4 of embodiment:
D1, import the lorry blind area picture after the correction of n frames, chooses t frames;
D2, extraction area-of-interest RIO, PCA dimensionality reductions are carried out to ROI region;
Histograms of oriented gradients HOG features in D3, extraction test sample, utilize the support vector machines after training point
Class device carries out the identification of blind area pedestrian;
D4, discriminate whether to detect pedestrian;
If D5, it is unidentified arrive pedestrian, choose latter two field picture, carry out D2;
If D6, recognize pedestrian, tracking box (groundtruth) four corner location information are extracted;
D7, according to tracking box (groundtruth) positional information, the color characteristic training of target in frame is carried out, before statistics
The color probability in scape target and background region;
D8, according to tracking box (groundtruth) positional information, the HOG features of detection pedestrian are simultaneously instructed in extraction frame
Practice, obtain associated filter template;
Image after D9, input next frame correction;
D10, using bayes method differentiate that each pixel in image belongs to the probability of prospect, then suppresses the similar face in edge
The object of color, obtains target following region of the present frame based on color characteristic;
D11, do Cosine Window processing, is then multiplied by Fast Fourier Transform (FFT) with associated filter template, is Fourier
Inverse transformation obtains peak response point diagram, obtains the target following region that present frame is obtained based on correlation filtering;
D12, the peak response point diagram progress for obtaining the target following region counted based on color characteristic and correlation filtering
Fusion, obtains final tracking box (groundtruth) region of target in present frame, is then transferred to D7.
Claims (4)
1. a kind of load truck right-hand bend safety pre-warning system, including fish-eye camera (1), bridge are equipped with liquid crystal display
(5), it is characterised in that railway carriage or compartment back is equipped with fish-eye camera (1) after truck cab, is equipped with the right side of boxcar more than one
Ultrasonic sensor (2), cabin interior are equipped with signal processing unit (3) and Central Control Center (4);Fish-eye camera (1)
Output terminal the input terminals of signal processing unit (3) is connected with ultrasonic sensor (2) output terminal, signal processing unit (3)
Output terminal connection Central Control Center (4) input terminal and liquid crystal display (5) input terminal, Central Control Center (4) it is defeated
Input terminal, the input terminal of ultrasonic sensor (2) and the input terminal of throttle control (6) of outlet connection fish-eye camera (1).
2. a kind of load truck right-hand bend safe early warning method, it includes the following steps:
Step 1: load truck right-hand bend dead zone information gathers:
After the Central Control Center of load truck detects right turn signal, start the fish-eye camera at railway carriage or compartment back after truck cab
(1), real-time right-hand bend blind area video is obtained, while starts the ultrasonic sensor (2) on the right side of load truck compartment, obtains distance
Information;
Step 2: fish-eye camera (1) is demarcated using disk scaling board:
Because it is approximate with disk that fish eye lens shoots the photo come, accurate dotted line is provided using disk scaling board using one kind
Feature completes the calibration of fish-eye camera (1), and the inside and outside parameter and distortion that fish-eye camera (1) is obtained by demarcating are
Number;
Step 3: the correction of flake fault image:
Fault image is corrected using the template in fish eye images function correction method, and with cubic convolution method to correction after
Image be filled, do homework for the detection and tracking of pedestrian in right-hand bend blind area;
Step 4: load truck right-hand bend blind area pedestrian detection and tracking:
After correcting distorted image, the detection and tracking processing of pedestrian are carried out in signal processing unit (3), after choosing correction
The first two field picture carry out load truck right-hand bend blind area pedestrian detection;Character representation using histograms of oriented gradients HOG adds
The classifier algorithm of upper support vector machines carries out pedestrian detection;After detecting blind area pedestrian, the tracking box position of the first frame is extracted
Confidence ceases, and blind area pedestrian tracking is carried out using the correlation filtering blending algorithm based on color characteristic and HOG character representations;
Step 5: load truck right-hand bend speed controls:
Load truck right-hand bend blind area pedestrian information is shown on liquid crystal display (5) in real time can help driver's understanding blind
Area's information, but driver fault misstepping on accelerator in order to prevent, using a kind of load truck right-hand bend speed based on people's car lateral separation
Control program, in fish-eye camera (1) detecting and tracking blind area pedestrian at the same time, ultrasonic sensor (2) obtain pedestrian and lorry
Lateral separation, Central Control Center (4) is according to the lateral separation between current lorry turning speed and pedestrian, active control lorry
Throttle control (6), prevents driver's maloperation;The specific method of control is:
Under any lateral separation, when throttle increases suddenly, Central Control Center carries out emergency braking;
A、10m<For lorry to the lateral separation of pedestrian, Central Control Center control throttle, limits turning speed 30km/h;
B、5m<Lateral separation of the lorry to pedestrian<10m, Central Control Center control throttle, limitation turning speed 20km/h;
C、3m<Lateral separation of the lorry to pedestrian<5m, Central Control Center control throttle, limitation turning speed 10km/h;
D、1m<Lateral separation of the lorry to pedestrian<3m Central Control Center controls throttle, limitation turning speed 5km/h;
E, lateral separation of the lorry to pedestrian<1m, Central Control Center control throttle, limitation turning speed 0km/h;
Repeat the operation that above step one arrives step 5.
3. a kind of load truck right-hand bend safe early warning method according to claim 2, it is characterised in that in the step 2
Fish-eye camera is demarcated using disk scaling board, which specifically includes:
B1, using fish-eye camera shoot multiple chequered with black and white disk scaling board images;
B2, carry out binaryzation, and cromogram is converted into gray-scale map;
B3, input skew disk uncalibrated image quantity, i.e., the number in the center of circle in transverse and longitudinal coordinate;
B4, using lower-left circle disk center as image coordinate origin, extract the position (x of each circle disk center0,y0) it is used as characteristic point;
B5, do sub-pix precision, keeps these centers of circle;
B6, calculate the spin matrix R in outer parameter with world coordinate system transfer principle according to camera coordinates system and be translated towards
T is measured, intrinsic parameter f is calculated according to camera coordinates system and image physical coordinates system transfer principlex,fy,cx,cy;
B7, using the nonlinear model formula of camera calculate coefficient of radial distortion k1,k2,k3,k4And tangential distortion coefficient p1,
p2。
4. a kind of load truck right-hand bend safe early warning method according to claim 2, it is characterised in that to load truck
Right-hand bend blind area carries out pedestrian detection and tracking, the step specifically include:
D1, import the lorry blind area picture after the correction of n frames, chooses t frames;
D2, extraction area-of-interest RIO, PCA dimensionality reductions are carried out to ROI region;
Histograms of oriented gradients HOG features in D3, extraction test sample, utilize the support vector machines grader after training
Carry out the identification of blind area pedestrian;
D4, discriminate whether to detect pedestrian;
If D5, it is unidentified arrive pedestrian, choose latter two field picture, carry out D2;
If D6, recognize pedestrian, four corner location information of tracking box are extracted;
D7, according to tracking box positional information, carry out the color characteristic training of target in frame, count foreground target and background area
Color probability;
D8, according to tracking box positional information, the HOG features of detection pedestrian are simultaneously trained in extraction frame, obtain correlation filter
Template;
Image after D9, input next frame correction;
D10, using bayes method differentiate that each pixel in image belongs to the probability of prospect, then suppresses edge Similar color
Object, obtains target following region of the present frame based on color characteristic;
D11, do Cosine Window processing, is then multiplied by Fast Fourier Transform (FFT) with associated filter template, does Fourier's inversion
Change and obtain peak response point diagram, obtain the target following region that present frame is obtained based on correlation filtering;
D12, merged the peak response point diagram that the target following region counted based on color characteristic and correlation filtering are obtained,
The final tracking box region of target in present frame is obtained, is then transferred to D7.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201711112829.0A CN107985200B (en) | 2017-11-13 | 2017-11-13 | Right-turning safety early warning method for load-carrying truck |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201711112829.0A CN107985200B (en) | 2017-11-13 | 2017-11-13 | Right-turning safety early warning method for load-carrying truck |
Publications (2)
Publication Number | Publication Date |
---|---|
CN107985200A true CN107985200A (en) | 2018-05-04 |
CN107985200B CN107985200B (en) | 2020-11-13 |
Family
ID=62031323
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201711112829.0A Active CN107985200B (en) | 2017-11-13 | 2017-11-13 | Right-turning safety early warning method for load-carrying truck |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN107985200B (en) |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109109748A (en) * | 2018-10-08 | 2019-01-01 | 南京云计趟信息技术有限公司 | A kind of pedestrian's identification early warning system for blind area on the right side of heavy motor truck |
CN109345874A (en) * | 2018-12-07 | 2019-02-15 | 安徽江淮汽车集团股份有限公司 | A kind of vehicle right-hand rotation method for early warning and system based on V2X |
CN110228413A (en) * | 2019-06-10 | 2019-09-13 | 吉林大学 | Oversize vehicle avoids pedestrian from being involved in the safety pre-warning system under vehicle when turning |
CN110929606A (en) * | 2019-11-11 | 2020-03-27 | 浙江鸿泉车联网有限公司 | Vehicle blind area pedestrian monitoring method and device |
CN112466159A (en) * | 2020-11-30 | 2021-03-09 | 浙江科技学院 | Right-turning safety early warning system for large vehicle |
TWI763150B (en) * | 2020-11-26 | 2022-05-01 | 中華電信股份有限公司 | System and method for driving safety and accident judgment and computer-readable medium |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US6661449B1 (en) * | 1996-06-06 | 2003-12-09 | Fuji Jukogyo Kabushiki Kaisha | Object recognizing apparatus for vehicle and the method thereof |
CN102923000A (en) * | 2012-11-01 | 2013-02-13 | 西南交通大学 | Automobile active backing speed limiting control method based on binocular vision target detection |
US20130100287A1 (en) * | 2011-10-25 | 2013-04-25 | Altek Autotronics Corp. | Blind Spot Detection System and Blind Spot Detection Method Thereof |
CN105678787A (en) * | 2016-02-03 | 2016-06-15 | 西南交通大学 | Heavy-duty lorry driving barrier detection and tracking method based on binocular fisheye camera |
CN205601701U (en) * | 2016-03-21 | 2016-09-28 | 南京信息工程大学 | Freight train bend pedestrian detects warning system |
CN106379317A (en) * | 2016-09-09 | 2017-02-08 | 张晶 | Auxiliary blind area display system for automobile safety |
-
2017
- 2017-11-13 CN CN201711112829.0A patent/CN107985200B/en active Active
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US6661449B1 (en) * | 1996-06-06 | 2003-12-09 | Fuji Jukogyo Kabushiki Kaisha | Object recognizing apparatus for vehicle and the method thereof |
US20130100287A1 (en) * | 2011-10-25 | 2013-04-25 | Altek Autotronics Corp. | Blind Spot Detection System and Blind Spot Detection Method Thereof |
CN102923000A (en) * | 2012-11-01 | 2013-02-13 | 西南交通大学 | Automobile active backing speed limiting control method based on binocular vision target detection |
CN105678787A (en) * | 2016-02-03 | 2016-06-15 | 西南交通大学 | Heavy-duty lorry driving barrier detection and tracking method based on binocular fisheye camera |
CN205601701U (en) * | 2016-03-21 | 2016-09-28 | 南京信息工程大学 | Freight train bend pedestrian detects warning system |
CN106379317A (en) * | 2016-09-09 | 2017-02-08 | 张晶 | Auxiliary blind area display system for automobile safety |
Cited By (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109109748A (en) * | 2018-10-08 | 2019-01-01 | 南京云计趟信息技术有限公司 | A kind of pedestrian's identification early warning system for blind area on the right side of heavy motor truck |
CN109345874A (en) * | 2018-12-07 | 2019-02-15 | 安徽江淮汽车集团股份有限公司 | A kind of vehicle right-hand rotation method for early warning and system based on V2X |
CN110228413A (en) * | 2019-06-10 | 2019-09-13 | 吉林大学 | Oversize vehicle avoids pedestrian from being involved in the safety pre-warning system under vehicle when turning |
CN110228413B (en) * | 2019-06-10 | 2020-07-14 | 吉林大学 | Safety early warning system for avoiding pedestrians from being involved under large-scale vehicle during turning |
CN110929606A (en) * | 2019-11-11 | 2020-03-27 | 浙江鸿泉车联网有限公司 | Vehicle blind area pedestrian monitoring method and device |
TWI763150B (en) * | 2020-11-26 | 2022-05-01 | 中華電信股份有限公司 | System and method for driving safety and accident judgment and computer-readable medium |
CN112466159A (en) * | 2020-11-30 | 2021-03-09 | 浙江科技学院 | Right-turning safety early warning system for large vehicle |
CN112466159B (en) * | 2020-11-30 | 2022-05-06 | 浙江科技学院 | Right-turning safety early warning system for large vehicle |
Also Published As
Publication number | Publication date |
---|---|
CN107985200B (en) | 2020-11-13 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN107985200A (en) | A kind of load truck right-hand bend safety pre-warning system and method | |
CN108189838B (en) | A kind of pattern of fusion adaptive cruise curve control method and device | |
EP2352136B1 (en) | System for monitoring the area around a vehicle | |
CN107031623A (en) | A kind of road method for early warning based on vehicle-mounted blind area camera | |
CN109435852A (en) | A kind of panorama type DAS (Driver Assistant System) and method for large truck | |
CN109190523B (en) | Vehicle detection tracking early warning method based on vision | |
JP5529910B2 (en) | Vehicle periphery monitoring device | |
CN104751600B (en) | Anti-fatigue-driving safety means and its application method based on iris recognition | |
CN107133559B (en) | Mobile object detection method based on 360 degree of panoramas | |
CN106156725A (en) | A kind of method of work of the identification early warning system of pedestrian based on vehicle front and cyclist | |
CN203020198U (en) | Auxiliary anti-collision reversing device for forklift | |
CN201240344Y (en) | Embedded integrated vision auxiliary driving safety system | |
CN107229906A (en) | A kind of automobile overtaking's method for early warning based on units of variance model algorithm | |
CN109733283A (en) | Be blocked obstacle recognition early warning system and identification method for early warning based on AR | |
CN108944668A (en) | A kind of auxiliary driving warning method for looking around input based on vehicle-mounted 360 degree | |
JP2011076214A (en) | Obstacle detection device | |
CN105059184A (en) | Device for early warning, active prevention, and active control of side tumbling of passenger vehicle at bend, and judgment method thereof | |
CN106853794A (en) | Inner wheel difference warning method and system for sensing object distance by using structure light | |
CN103699899B (en) | Method for detecting lane lines based on equidistant curve model | |
Chang et al. | Design and implementation of an intelligent motorcycle helmet for large vehicle approach intimation | |
KR20120086577A (en) | Apparatus And Method Detecting Side Vehicle Using Camera | |
CN110929606A (en) | Vehicle blind area pedestrian monitoring method and device | |
CN105787456A (en) | Method for detecting pedestrians in night far infrared images | |
CN206141431U (en) | Car is parked and is assisted safety coefficient based on kinect | |
CN103077393B (en) | Based on vehicle-mounted real time kinematics object detection system and the method thereof of DSP |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |